Deep Brain Stimulation (DBS) of the thalamus, globus pallidus (GPi) or subthalamic nucleus (STN) is a new treatment for essential tremor and bradykinesia in Parkinson's Disease (PD). The neuropsychological and mood effects of stimulation at these sites are poorly categorized and understood. Major depression (MD) occurs in up to 40% of PD patients. The functional neuroanatomy of normal mood regulation in health and the pathological changes in MD are becoming better understood but still lag behind the neuroanatomical knowledge of motor dysfunction in PD. A key concept developed by the PI and others is that mood is regulated through changes in activity in the anterior paralimbic circuit (APLC) (amygdala, septum, anterior cingulate cortex, anterior temporal poles and orbitofrontal cortex) and that this system functions abnormally in MD. The PI and others have pioneered new treatments for MD JMS, VNS) that have indirectly stimulated this brain circuit. In modem research, no one has used DBS for the primary treatment of MD, although mood effects of DBS have been observed. Based on functional imaging, case report and animal model data, we hypothesize that DBS will cause acute mood changes in PD patients only when DBS affects the APLC. To test this theory that APLC activation is necessary for mood-altering and perhaps antidepressant effects of DBS, we propose to recruit 32 PD patients over a three year period who would fulfill PD clinical indications for DBS (e.g. resistant akinesia, postural rigidity). We will randomly assign these subjects to bilateral DBS at one of two sites either the STN or the GPi. Two days following implantation we will use interleaved DBS/fMRI and intermittently stimulate at the 4 sites on each electrode (bilaterally coordinated), examining changes in self rated mood, psychophysiological measures (HR, BP, GSR, Oxygen saturation), and most importantly, rCBF changes locally and in secondary limbic regions (APLQ. In a masked clinical trial, we will then examine the effects of 4 weeks of a fixed dose of clinically indicated DBS at the implanted site (GPi, STN) on movement, mood and cognition. Subjects with depression symptoms following the 'standard' care will be treated for 4 weeks at the electrode site that on fMRI had the most APLC involvement. The goals of this study are to test this APLC hypothesis of mood regulation and to develop a better working knowledge of the effects of DBS on mood in PD. Thus, we plan to use interleaved fMRI and DBS to directly examine the effects of DBS on hypothesized brain circuits and to correlate this with immediate and longer-term (4 week) behavioral outcomes (mood and cognition).